Title :
On the selection of median structure for image filtering
Author :
Reeves, Stanley J.
Author_Institution :
Dept. of Electr. Eng., Auburn Univ., AL, USA
fDate :
8/1/1995 12:00:00 AM
Abstract :
Median filtering is a powerful tool for reducing noise in images, particular for long-tailed noise distributions. However, the choice of filter mask is critical. The proper choice depends on the image and noise statistics, which are often unknown. We propose cross-validation as a method for selecting a median filter structure directly from the corrupted image data. This method requires no knowledge of the statistics of the noise or image. We demonstrate the value of this method with several examples
Keywords :
filtering theory; image enhancement; median filters; corrupted image data; cross-validation; filter mask; image filtering; long-tailed noise distributions; median filter structure; median structure selection; noise reduction; Additive noise; Data analysis; Filtering; Filters; Image restoration; Laplace equations; Maximum likelihood estimation; Noise reduction; Statistical distributions; Statistics;
Journal_Title :
Circuits and Systems II: Analog and Digital Signal Processing, IEEE Transactions on
